Investigation of incipient dynamic stall over pitching airfoils at high Reynolds numbers

Surface shear stress and pressure measurements were obtained on a NACA 0012 airfoil model undergoing a pitch-up motion from 0° to 45° angle of attack at a constant rate. The shear-stress data were obtained using an array of surface-mounted hot-film sensors. Dominant features in these data and in the standard deviations computed from these data were examined and related to events in the development and evolution of the dynamic stall vortex over the suction surface. These features were compared with flow visualization results from prior studies and well-known features of the dynamic stall process seen in the surface-pressure distributions. Trends in the behavior of these features are presented for a range of non-dimensional pitch rates (0.010 < α + < 0.150) and chord Reynolds numbers (0.9 × 105 < Re_c < 1.74 × 106). In general, at high Reynolds numbers, the events that lead to the formation of the dynamic stall vortex are delayed to higher angles of attack as the pitch rate is increased. This is consistent with prior work performed at low Reynolds numbers. However, significant changes were seen in the behavior of the flow features at high Reynolds numbers. Transition to turbulence in the boundary layer over the airfoil suction surface early in the pitch-up motion significantly modifies the unsteady separation process. The point of transition appears as a sudden increase in wall shear stress accompanied by a distinct peak in the standard deviation of the hot-film sensor output. At low pitch rates, the location and angle of attack of transition in the leading-edge region remain independent of Reynolds number. At high pitch rates, however, there is a significant delay in angle of attack of transition at higher Reynolds numbers. At high pitch rates, dynamic stall vortex formation is delayed to higher angles of attack with increasing Reynolds number. At low pitch rates, however, this delay is noticeably reduced at higher Reynolds numbers. This behavior is due to both the quasi-steady nature of the flow separation in these cases and the effects of compressibility. It is believed that compressibility effects play a significant role at the lower pitch rates and higher Reynolds numbers investigated in this study, since high local Mach numbers were seen in the leading edge region for these cases. Additional experiments are required to separate the effects of Reynolds number and Mach number. Also, since transition noticeably effected all cases investigated in this study, experiments should be performed over a wider range of pitch rates and Reynolds numbers in order to pinpoint the critical Reynolds numbers at each pitch rate where transition first begins to modify the unsteady separation process.